1. Field of the Invention
The present invention is directed to a method and to an apparatus for generating a volume dataset of a subject with a C-arm X-ray apparatus while taking a movement of the subject during a registration of a series of 2D projections of the subject into consideration.
2. Description of the Prior Art
For generating a volume dataset of a subject, a C-arm of a C-arm X-ray apparatus is rotated through approximately 190xc2x0 around the subject in a time span of several seconds, with the 2D projections of the series being registered from different projection directions of the subject. The volume dataset of the subject can be generated from the 2D projections using projection matrices in a reconstruction algorithm.
Published United States Application 2001/0053204 discloses a calibration method for a C-arm X-ray apparatus, wherein projection geometries of the C-arm X-ray apparatus that are stored in projection matrices are determined with the assistance of a position acquisition system in order to be able to reconstruct a volume dataset of the subject in a subject measurement from a series of 2D projections acquired from different projection directions.
U.S. Pat. No. 6,289,235 discloses a method wherein markers are arranged in a fixed position relative to a subject. 2D projections of the subject and the markers are acquired from different projection directions with a radiation source and a reception device. Projection geometries for the reconstruction of a 3D image of the subject can be derived from the markers imaged in the 2D projections.
Movements on the part of the subject to be imaged, for example respiratory movements of a patient, during the registration of the 2D projections are problematical for the generation of the volume dataset. These respiration-produced movements of a patient are not compensated by the reconstruction algorithm. Movements of a subject during the registration of a series of 2D projections accordingly lead to a qualitatively poorer result of the reconstruction, i.e. to a qualitatively poorer volume dataset of the subject. In medical applications, the volume dataset generated from a patient even can be diagnostically useless.
In order to counteract respiration-produced reconstruction artifacts, images of patients are registered with ECG-triggering. ECG-triggering, however, has the disadvantage that additional apparatus are required for generating a volume dataset, which can be an impediment particularly for intra-operative employment.
German OS 41 37 652 discloses a method wherein an X-ray CT, scanner acquires projection data from a series of views during the scanning of a rib cage of a patient. Movements of the rib cage of the patient due to respiration during the scanning are thereby also acquired. The motion data are employed together with a geometrical model of the rib cage movement in order to calculate factors that correct the acquired projection data and reduce motion artifacts in an image.
An object of the present invention is to provide a method and an apparatus of the type described above wherein the generation of a volume dataset of a moving subject is simplified.
This object is inventively achieved in a method and an apparatus for generating a volume dataset of a moving subject, wherein a referencing base is arranged at the subject in a defined way, and the position and the orientation of the referencing base are acquired by a position acquisition system. During the exposure of the series of 2D projections of the subject, the position and the orientation of the referencing base is determined with the position acquisition system and the position of the C-arm is determined substantially simultaneously with at least one exposure of a 2D projection of the series of 2D projections. Moreover, the deviation of the current position and of the current orientation of the referencing base from a previously defined reference position of the referencing base is determined for the 2D projection. Via the referencing base arranged at the subject in a defined way, accordingly, the deviations of the current position and of the current orientation of the subject from its reference position, corresponding with the reference position of the referencing base can be determined. Since the deviation of the referencing base or the deviation of the subject from its reference position has been determined for a 2D projection and the position of the C-arm for the 2D projection also has been determined, these can be taken into consideration in the generation of the volume dataset, so that a qualitatively higher-grade volume dataset can be generated.
In a version of the invention, the imaging of the subject in the 2D projection is shiftedxe2x80x94before the generation of the volume datasetxe2x80x94on the identified deviation of the referencing base allocated to this 2D projection or, the deviation of the subject from its reference position, and the position of the C-arm allocated to the 2D projection, for taking a movement of the subject in at least one 2D projection of the series of 2D projections into consideration. In this way, a 2D projection is generated with an imaging of the subject shifted according to the deviation of the position and the orientation of the subject from its reference position, and thus the object movement is compensated. A translational shift of the imaging of the subject parallel to the image plane of the 2D projection preferably ensues, particularly since subject movements perpendicular to the image plane of the 2D projection have only a minimal influence on the 2D projection and the reconstruction result. As a rule, the imaging of the subject is shifted in all 2D projections of the series of 2D projections corresponding to its deviation from its reference positions, and the reconstruction algorithm for generating the volume dataset is implemented only subsequent thereto. Because practically all movements of the subject during the exposure of the series of 2D projections are taken into consideration by the corresponding shifts of the images of the subject in the 2D projections, a qualitatively higher-grade volume dataset of the subject is obtained.
In another embodiment of the invention each projection matrix is modifiedxe2x80x94before the generation of the volume datasetxe2x80x94based on the deviation of the identified referencing base allocated to its 2D projection, or on the basis of the identified deviation of the subject from its reference position, and the position of the C-arm allocated to this 2D projection for taking a movement of the subject into consideration for at least one 2D projection of the series of 2D projections. In a second version of this embodiment, at least one extrinsic imaging parameter of the projection matrix is modified. Whereas the movements of the object during the exposure of the 2D projections in the above-described version are compensated by shifts of the images of the subject in the 2D projections, the second version of the invention is based on the consideration of viewing the subject as a system at rest and viewing the X-ray system as a moving system. In the case of the second version, accordingly, the extrinsic imaging parameters of the projection matrices are modified according to the movements of the subject since the extrinsic imaging parameters describe the position and orientation of the X-ray system, for example the focus of the X-ray source. By suitable modification of the extrinsic imaging parameters of the projection matrices, accordingly, movements of the subject during the exposure of the series of 2D projections can likewise be taken into consideration. When generating the volume dataset of the subject, only the modified projection matrices are employed by the reconstruction algorithm, so that a higher-quality volume dataset again is achieved.
In another embodiment of the invention the reference position of the referencing base, which corresponds with the reference position of the subject, is fixed with the exposure of the first 2D projection of the series of 2D projections.